Note: Descriptions are shown in the official language in which they were submitted.
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
TARGETED WEED CONTROL WITH CHEMICAL AND MECHANICAL MEANS
FIELD OF THE INVENTION
This invention relates generally to weed management, and more specifically to
a method for
weed management in a plantation field. The invention relates further to a
decision-support
system in a plantation field, to a treatment device for weed treatment in a
plantation field, and to
a system for weed management in a plantation field.
BACKGROUND OF THE INVENTION
Commercial farms often adopt chemical control using selective or non-selective
herbicides.
However, the weed management gets increasingly under pressure, as resistance
may develop
when a target adapts to circumvent a particular control strategy. For example,
herbicide
resistance may be inherited and may occur through selection of weeds that
survive herbicide
treatment. With repeated selection, resistant weeds may multiply until they
dominate the
population. If weed management is not done properly, resistance may develop in
any field, as
resistance may have been influenced by proximity to another affected field and
farm gateways
on the farm. Herbicide strategy may be optimized to ensure that resistance
does not develop to
these herbicide by paying attention to recommended rate, nozzle choice, water
volumes and
spray timing to maximize product effectiveness.
SUMMARY OF THE INVENTION
There may be a need to improve weed management in a plantation field.
The object of the present invention is solved by the subject-matter of the
independent claims.
Further embodiments and advantages of the invention are incorporated in the
dependent
claims. The described embodiments similarly pertain to the method for weed
management in a
plantation field, to the decision-support system in a plantation field, to the
treatment device for
weed treatment in a plantation field, and to the system for weed management in
a plantation
field.
A first aspect of the invention relates to a method for weed management in a
plantation field.
The method comprises collecting image data of the plantation field after being
applied with a
primary weed control technique, recognizing items indicative of an occurrence
of a weed in the
collected image data, generating a weed pressure map indicative of a weed
pressure in the
plantation field based on the recognized items, determining one or a plurality
of target areas for
1
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
weed management in the generated weed pressure map, if the weed pressure of
the one or the
plurality of target areas extends a predetermined threshold, and outputting a
treatment control
signal indicative of the one or the plurality of determined target areas,
which when transmitted
causes an activation of at least one weed treatment device to apply at least
one secondary
weed control technique to the one or the plurality of determined target areas,
wherein the at
least one secondary weed control technique is different from the primary weed
control
technique.
In other words, the primary weed control technique is applied to the
plantation field at a field
scale. The primary weed control technique may be less expensive and/or non-
selective, such
that the majority of the weed types in the plantation field can be eradicated
with relatively low
costs. However, in some situations the primary weed control technique cannot
eradicate all
weed species. For example, there may be an area with weeds that are tolerant
to the primary
weed control technique. For example, if the primary weed control technique
uses herbicides,
there may be an area with herbicide-resistant weeds. Thus, in addition to the
primary weed
control technique applied at a field scale, the secondary weed control
technique may be
advantageously applied to the field at a local scale targeting at areas that
requires an additional,
different treatment. The combination of the application of the primary weed
control technique at
a field scale and the secondary weed control technique at a local level may
allow an efficient
and effective weed management in the plantation field.
According to an embodiment of the invention, the method further comprises
applying, with the at
least one weed treatment device, at least one secondary weed control technique
to the one or
the plurality of determined target areas based on the treatment control
signal.
For example, the treatment control signal may indicate the geographical
information of the
target areas. The geographical information may be used to provide guidance of
GPS-guided
ground robots, or aerial sprayers to apply weed control only at these target
areas.
According to an embodiment of the invention, the primary weed control
technique is a chemical
weed control technique with a first active ingredient. The at least one
secondary weed control
technique comprises a chemical weed control technique with a second active
ingredient and/or
a mechanical weed control technique.
The first active ingredient may be less expensive than the second active
ingredient. The first
active ingredient may be non-selective, whereas the second active ingredient
may be selective
for specific weed species. The mechanical weed control technique may be used
e.g. when the
2
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
target areas comprise herbicide-resistant weeds. In other words, integrated
procedures may be
adopted that involves a few different methods for a better eradication of
weeds in the target
areas.
According to an embodiment of the invention, the determination of one or a
plurality of target
areas comprises the step of determining a geographical information of the one
or the plurality of
target areas.
For example, sensors may allow each and every image to be tagged with GPS
coordinates. The
GPS coordinates may be used to provide guidance of GPS-guided ground robots,
or aerial
sprayers to apply weed control only at these target areas.
According to an embodiment of the invention, the determination of one or a
plurality of target
areas comprises the step of recognizing a single or a plurality of weeds in
the one or the
plurality of determined target areas for weed management and the step of
determining a weed
type and/or a weed density of the single or the plurality of recognized weeds.
An automated machine-based determination or recognition process may be used
for a specific
type of weed starting from the image data, pre-processing of the image data,
deriving metadata
.. from the digital image and use there by e.g. neural network based
classifier for a probability
based analysis of the image data, and finally a recognition on one or more
types of weeds.
Spectral signatures, thermal signatures and/or texture specifications of the
weeds may be used
for distinguishing weeds and crops depending on the types of image data. This
may allow
choosing appropriate secondary weed control technique for the specific weed
and/or intensity.
According to an embodiment of the invention, the determination of one or a
plurality of target
areas comprises identifying rows and/or a density of a single or a plurality
of crops in the
plantation field based on the collected image data.
.. For example, spectral imagery or thermal imagery may be used to distinguish
the spectral
signatures of weed and crops and to identify the rows of crops in the
plantation field. This may
advantageously allow a better identification of the target areas for weed
treatment.
According to an embodiment of the invention, the determination of one or a
plurality of target
.. areas comprises the step of identifying a single or a plurality of target
areas with at least one
resistant weed based on historical information of the plantation field.
3
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
Weed infestation maps of previous seasons may be used to determine these
target areas. As
these target areas comprise herbicide-resistant weeds, the secondary weed
control technique
may be other weed control technique, such as mechanical weed control
techniques, for a better
weed control.
According to an embodiment of the invention, the determination of one or a
plurality of target
areas for weed management may further comprise the step of identifying a
single or a plurality
of target areas, where the use of crop protection products is not recommended,
based on
historical information of the plantation field and recommending an application
of a mechanical
weed control technique in the single or the plurality of the identified target
areas.
For example, weed infestation maps of previous seasons may be used, which
provides the
information about areas where the use of crop protection products is not
effective or not
preferred. In these areas, mechanical weed control techniques may be applied
for a better weed
management.
According to an embodiment of the invention, the determination of one or a
plurality of target
areas for weed management may further comprise determining, for each
determined target
area, a respective secondary weed control technique. The treatment control
signal is indicative
of the respective determined secondary weed control technique for each target
area, which
when transmitted causes an activation of at least one weed treatment device to
apply, for each
determined target area, the respective determined secondary weed control
technique.
The respective secondary weed control technique may be determined based on at
least one of
the following: i) the generated weed pressure map, ii) the determined
geographical information
of the single or the plurality of target areas with a weed pressure above a
predefined threshold,
iii) the determined weed type and/or the determined weed density of the single
or the plurality of
target areas, iv) the identified rows and/or the identified density of the
single or the plurality of
crops in the plantation field, v) the identified single or the identified
plurality of target areas with
at least one resistant weed, and vi)the identified one or plurality of target
areas to be applied
with a mechanical weed control technique. This may advantageously allow a
choice of an
appropriate weed control technique depending on the weed species, intensity
and/or historical
information of the plantation field and to apply an appropriate weed control
technique to which
the weeds are susceptible.
4
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
According to an embodiment of the invention, the application of at least one
secondary weed
control technique comprises applying, for each determined target area, the
respective
determined secondary weed control technique based on the treatment control
signal.
.. The secondary weed control technique and the target areas may be provided
as configuration
information, which may be loaded onto a respective weed treatment device to
regulate the
activity of weed treatment devices.
According to an embodiment of the invention, the method further comprises the
step of
generating a field record. The field record comprises at least one of the
following: i) the
generated weed pressure map, ii) the determined geographical information of
the single or the
plurality of target areas with a weed pressure above a predefined threshold,
iii) the determined
weed type and/or the determined weed density of the single or the plurality of
target areas, iv)
the identified rows and/or the identified density of the single or the
plurality of crops in the
.. plantation field, v) the identified single or the identified plurality of
target areas with at least one
resistant weed, and vi) the identified one or plurality of target areas to be
applied with a
mechanical weed control technique.
The field report may be used for online or offline precision farming. The
field report may provide
a detailed information of occurrence and spread of weed species in the
plantation field.
Additionally, the field record of previous seasons may also provide a detailed
information of
areas, which are prone to weed infestation.
A second aspect of the invention relates to a decision-support system for weed
management.
The decision-support system comprises a data interface, an image recognition
unit, a weed
pressure mapping unit, a determination unit, a controlling unit, and a
treatment control interface.
The image recognition unit is configured to recognize items indicative of an
occurrence of a weed
in image data of the plantation field received from the data interface. The
weed pressure mapping
unit is configured to generate a weed pressure map indicative of a weed
pressure in the plantation
.. field based on the recognized items. The determination unit is configured
to determine one or a
plurality of target areas for weed management in the generated weed pressure
map, if the weed
pressure of the one or the plurality of target areas extends a predetermined
threshold. The
controlling unit is configured to generate a treatment control signal
indicative of the one or the
plurality of determined target areas and to output the treatment control
signal to the treatment
control interface, which when transmitted causes an activation of at least one
weed treatment
device to apply at least one secondary weed control technique to the one or
the plurality of
determined target areas.
5
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
In other words, a decision-support system may be used to detect occurrence and
map the
spread of weed species in a plantation field based on the collected image
data, e.g. spectral
data or thermal data, of the plantation field. The weed map produced may offer
a robust basis
for farmers to prepare weed control schedules, e.g. herbicide spray schedules,
and a
generalized weed control program and to regulate the activity of weed
treatment devices
accordingly. As will be explained hereafter, the decision-support system may
also be configured
to identify weeds, discriminate different types of weeds, and/or calculate
herbicide
requirements.
According to an embodiment of the invention, the determination unit is
configured to determine,
for each determined target area, a respective secondary weed control
technique. The treatment
control signal is indicative of the respective determined secondary weed
control technique for
each target area, which when transmitted causes an activation of at least one
weed treatment
device to apply, for each determined target area, the respective determined
secondary weed
control technique.
In other words, the decision-support system may control multiple treatment
devices, each
provided with a respective weed control technique and/or with a respective
herbicide, to apply
weed control only at targeted spots infested with weeds that are susceptible
to the respective
weed control technique and/or to the respective herbicide. Such an effort may
reduce herbicide
requirements and improve weed control efficiency. This may also reduce
contamination of
irrigation channels and ground water with herbicide.
A third aspect of the invention relates to a weed treatment device for weed
treatment in a
plantation field. The weed treatment device comprises a treatment control
interface and a
treatment arrangement with one or a plurality of treatment units. The
treatment control interface
of the weed treatment device is connectable to a treatment control of a
decision-support system
as described above and below and configured to receive a treatment control
signal indicative of
one or a plurality of target areas for weed treatment. The weed treatment
device is configured to
activate respective ones of treatment units of the treatment arrangement to
apply a secondary
weed control technique to the one or the plurality of target areas based on
the received
treatment control signal.
The weed treatment device may comprise ground robots with variable-rate
applicators, aerial
sprayers, or other variable-rate herbicide applicators to apply chemical weed
control
procedures. The weed treatment device may comprise ground robots with physical
means to
6
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
remove, injure, kill, or make the growing conditions unfavorable for the
weeds. Geographical
information may be used to guide various type of robots and sprayers to apply
appropriate weed
control technique only at spots infested with weeds and/or prone to weed
infestation.
According to an embodiment of the invention, the treatment control signal is
indicative of a
respective secondary weed control technique for each target area. The weed
treatment device
is configured to apply the secondary weed control technique to a target area
where the
indicated secondary weed control technique matches that of the weed treatment
device.
In other words, weed control methods may vary from one target area to another
for a better
weed control of the weeds in each area.
According to an embodiment of the invention, the weed treatment device is at
least one
selected from a mechanical weed treatment device configured to be steered by
the controlling
unit to apply a mechanical weed control technique to the one or the plurality
of target areas, and
a chemical weed treatment device configured to apply a second crop protection
product to the
one or the plurality of target areas.
This may advantageously allow weed control in a plantation field by adopting
integrated
procedures that involves at least two different methods. This may be
beneficial for the situations
where herbicides are not recommended and thus a mechanical weed control
technique is more
appropriate.
A fourth aspect of the invention relates to a system for weed management in a
plantation field.
The system comprises a sensor carrier, a decision-support system as described
above and
below, and at least one weed treatment device as described above and below.
The sensor
carrier is configured to collect image data of the plantation field after
being applied with a
primary weed control technique. The decision-support system is configured to
determine one or
a plurality of areas for weed management, and preferably, for each determined
target area, a
respective secondary weed control technique based on the collected image data,
and to control
the at least one weed treatment device to apply at least one secondary weed
control technique
to the one or the plurality of target areas.
The system may advantageously allow a weed control ranging from mission
planning, acquiring
imagery of a plantation field, spectral imaging processing, locating weeds,
noting the intensity of
infestation and mapping them, to practicing precision weed control techniques
based on the
weed infestation maps of this season optionally in combination with the
previous seasons.
7
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
BRIEF DESCRIPTION OF THE DRAWINGS
Exemplary embodiments of the invention will be described in the following with
reference to the
.. following drawings:
Fig. 1 shows a schematic drawing of a method according to an exemplary
embodiment of the
present disclosure.
Fig. 2 shows a schematic drawing of a decision-support system according to an
exemplary
embodiment of the present disclosure.
Fig. 3 shows a schematic drawing of a weed treatment device according to an
exemplary
embodiment of the present disclosure.
Fig. 4 shows a schematic drawing of a plantation field according to an
exemplary embodiment
of the present disclosure.
Fig. 5 shows a schematic drawing of a system for weed treatment according to
an exemplary
embodiment of the present disclosure.
DETAILED DESCRIPTION OF DRAWINGS
In the context of this description, the following conventions, terms and/or
expressions may be
used:
The term "image data" as used herein may denote image data captured with
visual, infrared
(IR), near infrared (N IR), multispectral and/or thermal sensors or cameras.
The wavelength
range of the sensor or camera may be selected for a better judgment of a
specific weed type.
For example, cameras with facility for color-infrared (CIR), thermal infrared
(TIR), and visible
N IR (VN IR) bandwidth imagery could be adopted to judge the effect of
weedicide glyphosate.
The image data may be collected using ground, satellite or drone platforms.
The term "items indicative of an occurrence of a weed" as used herein may
denote a spectral
signature of a weed, i.e. the particular spectral response of the weed in the
image data that is
different from the crops. The image data may be captured with visual, IR, N
IR, multispectral
sensors. The items indicative of an occurrence of a weed may also denote the
thermal
8
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
response of the weed that is different from the crops in the image data
captured with a thermal
sensor. The items indicative of an occurrence of a weed may also denote color
or texture
specification of a weed that is different from the crops. The texture
specification may relate to
leaf veins, characteristic forms of leaf segments, specific patterning and
color distribution, micro
hairs on the surface atop of the leaf and the edge of the leaf.
The term "weed" as used herein may denote an unwanted plant of any species
that can quickly
adapt to any environment. Hence, the expression weed may also be related to
plant among
crop or cultivated or economic plants that are useful in the sense of
harvesting fruits of grown
up seed. Weed may disturb the process of the growth of the crop and decrease
the yield of an
agriculture field.
The term "weed pressure" (WP) as used herein may denote that an estimate of
the percentage
of weeds that contribute to the total cover of both crops and weeds in a given
area. Weed
pressure values may be ranged between 0 in case of total absence of weeds, and
100 in case
of total absence of crops. For simplify expert assessment of digital images,
four categories for
the weed pressure may be formulated. The interval bounds for the weed pressure
may be as
follows: WP1 (0-10), WP2 (11-20), WP3 (21-35), and WP4 (greater than 35).
.. The term "weed pressure map" as used herein may denote the weed pressure
distribution in a
plantation field. For example, the plantation field may be divided into a
plurality of grids in form
of a rectangular array of squares or rectangles of equal size. The weed
pressure may be
evaluated for each grid for generating a weed pressure map for the plantation
field.
The term "primary weed control technique" as used herein may denote a weed
control
technique that is applied to manage weeds at a field scale. For example, the
primary weed
control technique may be a weed control technique with lower weed-control
costs. As a further
example, the primary weed control technique may be a non-selective weed
control technique,
such as a single herbicide used to control most weeds - grasses, sedges and
broadleaves.
The term "secondary weed control technique" as used herein may denote a weed
control
technique that is applied to manage weeds at a local scale in the field. The
secondary weed
control technique may be a weed control technique with higher weed-control
costs. The
secondary weed control technique may be a selective weed control technique for
specific weed
types.
9
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
The term "weed control technique" as used herein may be grouped broadly into
mechanical,
chemical, and biological methods. Weed control techniques vary depending of
several factors
related to geographic location, cropping system followed, current crop
species, weed species
traced and their intensity, and economic consideration. Chemical control may
use selective or
non-selective herbicides based on weed flora. Mechanical weed control is any
physical activity
that inhibits unwanted plant growth.
The term "weed treatment device" as used herein may denote a chemical weed
treatment
device. For example, ground robots with variable-rate applicators, aerial
sprayers, or other
variable-rate herbicide applicators. The weed treatment device may denote
physical/mechanical
weed treatment device, e.g. ground robots using physical methods to remove,
injure, kill, or
make the growing conditions unfavorable for the weeds. The weed treatment
device may be
regulated by the configuration information. As will be explained in detail
hereafter, the
configuration may be generated by a decision-support system, e.g. a remote
server of a service
provider. The generated configuration information may be loaded onto the weed
treatment
device to enable the weed treatment device to perform weed treatment.
The term "decision-support system" as used herein may denote a computing
device or a
computing system, regardless of the platform, being suitable for executing
program code related
to the proposed method. For example, the decision-support system may be a
smartphone or
other mobile devices with a processor suitable for implementation of the
proposed method
and/or related decision-support system. As a further example, the decision-
support system may
also be a computer system. The decision-support system may also be a remote
server that
provides a web service to facilitate weed management of a plantation field
e.g. by a farmer of
the plantation field. The remote server may have a more powerful computing
power to provide
the service to multiple users to manage many different plantation fields. The
remote server may
include an interface through which a user can authenticate (e.g. by providing
a username and
password); and an interface for creating, modifying, and deleting
configuration information of the
weed treatment devices of the user. The configuration information may be
generated by the
decision-system by analyzing the image data. For example, the configuration
information may
comprise geographical information of the target area, start time for applying
the weed control,
active ingredient to be applied, etc. The configuration information may be
loaded onto the weed
treatment devices to enable the weed treatment devices to perform weed
treatment. It is also
noted that the image recognition unit, the weed pressure mapping unit, the
determination unit,
and the controlling unit of the decision-support system may be separate
devices. For example,
the controlling unit may be a smartphone with a display for controlling the
weed treatment
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
devices, whereas the image recognition unit, the weed pressure mapping unit,
the
determination unit are implemented with a computer.
Fig. 1 shows a block diagram of an embodiment of a method for weed management
in a
plantation field. In step S10, image data of the plantation field may be
collected after being
applied with a primary weed control technique. The primary weed control
technique may be e.g.
a weed control technique with lower weed-control costs and/or a non-selective
weed control
technique. The image data may be collected using ground, satellite or drone
platforms. To
collect the image data, ground robots, satellite or drones may be fitted with
visual, IR, N I R,
and/or thermal sensors. These sensors may provide spectral data and/or thermal
data of the
plantation field. The image data may be transmitted to a decision-support
system automatically
e.g. via a network, for further analysis. The image data may also be uploaded
by a user, e.g. via
a network, to a decision-support system for further analysis.
In step S20, items indicative an occurrence of a weed may be recognized in the
collected image
data. For example, spectral imagery may provide distinction among soil
surface, weeds in the
inter-row and crop in the rows, as weeds and crops have different spectral
responses.
In step S30, a weed pressure map indicative of a weed pressure in the
plantation field is
prepared based on the recognized items. For example, the plantation field may
be divided into a
plurality of grids in form of a rectangular array of squares or rectangles of
equal size. The weed
pressure may be evaluated for each grid for generating a weed pressure map for
the plantation
field.
In step S40, one or a plurality of target areas for weed management is
determined in the
generated weed pressure map, if the weed pressure of the one or the plurality
of target areas
extends a predetermined threshold. A treatment control signal is outputted
indicative of the one
or the plurality of determined target areas, which when transmitted causes an
activation of at
least one weed treatment device to apply at least one secondary weed control
technique to the
one or the plurality of determined target areas, wherein the at least one
secondary weed control
technique is different from the primary weed control technique. For example,
weed pressure
values may be ranged between 0 in case of total absence of weeds, and 100 in
case of total
absence of crops. For identification of the target areas, four categories for
the weed pressure
may be formulated. The interval bounds for the weed pressure may be as
follows: WP1 (0-10),
WP2 (11-20), WP3 (21-35), and WP4 (greater than 35). Areas with a weed
pressure value
falling into WP4 may, for example, be determined as target areas.
11
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
In optional step S50, at least one secondary weed control technique is applied
with the at least
one weed treatment device to the one or the plurality of target areas based on
the treatment
control signal. The at least one secondary weed control technique is different
from the primary
weed control technique. The secondary weed control technique may comprise a
chemical
control technique using e.g. expensive or selective herbicides based on weed
flora. The
secondary weed control technique may comprise, alternatively or additionally,
a mechanical
weed control technique. In other words, the secondary weed control technique
may adopt
integrated procedures that may involve a few different methods, such as a
combination of
mechanical and chemical procedures to eradicate weeds at a local scale.
In other words, the weed management at a field scale with the primary weed
control technique
is combined with the weed management at a local scale with at least one
secondary weed
control technique, for example, targeting at weeds that are resistant to the
primary weed control
technique but susceptible to the secondary weed control technique. As the
primary weed control
.. technique may be less expensive than the secondary weed control technique,
this may be
helpful in reducing costs on weed control. Additionally, the secondary weed
control technique
may be selected to be more effective for weeds that are resistant to the
primary weed control
technique, it is devoid of ineffective problem of the primary weed control
technique.
The primary and secondary weed control techniques may be selected based on
several factors,
such as geographic location, cropping system followed, current crop species,
weed species and
their intensity, and economic consideration. The weed may be identified using
their spectral
signatures. Often, the spectral signatures of weeds common to an area or
cropping system are
available. This may allow the discrimination between weed and crop using
spectral signatures
and further the judgment of the percentage of weeds that contribute to the
total cover of both
crops and weeds in a given area, i.e. weed pressure.
For example, the primary weed control technique is a chemical weed control
technique with a
first active ingredient. The at least one secondary weed control technique may
comprise a
chemical weed control technique with a second active ingredient and/or a
mechanical weed
control technique. The first active ingredient may be less expensive than the
secondary active
ingredient. Alternatively or additionally, the first active ingredient may be
non-selective, whereas
the second active ingredient may be selective for weeds that are resistant to
the first active
ingredient.
Optionally, the determination of one or a plurality of target areas for weed
management may
further comprise the step of determining S41 a geographical information of the
one or the
12
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
plurality of target areas. The geographical information may use Global
Position System (GPS)
for precision guidance of treatment devices, e.g. ground robots with e.g.
variable-rate applicator,
aerial sprayer, and/or other variable rate herbicide applicators or mechanical
methods.
Optionally, the determination of one or a plurality of target areas for weed
management may
further comprise the step of identifying S42 a single or a plurality of weeds
in the one or the
plurality of identified target areas for weed management, and determining S43
a weed type
and/or an intensity of the single or the plurality of recognized weeds. Target
weed species with
specific spectral reflectance pattern may be picked up by the multispectral
imagery. Weeds may
be identified and mapped using object-based image analysis (OBIA) using
segmentation. Weed
traits such as location, proximity and hierarchical relationships with the
spectra of the object
(weed/crop) may be used. A library or data bank of spectral signatures of
various weed species
may be used to discriminating different types of weeds. This may be done by
using a machine
learning technique. Machine learning algorithms build a mathematical model of
sample data of a
plurality of weeds (e.g. spectral data, or image data) known as "training
data", in order to make
predictions or decisions without being explicitly programmed to perform the
task. This may
advantageously allow the site-specific weed management and the selection of
appropriate
secondary weed control technique for target areas with different types of
weeds.
Optionally, the determination of one or a plurality of target areas for weed
management may
further comprise the step of determining S44 rows and/or a density of a single
or a plurality of
crops in the plantation field based on the collected image data. Crops may be
identified based
on specific spectral reflectance pattern. Similar to weed identification, crop
traits such as
location, proximity and hierarchical relationships with the spectra of the
object (weed/crop) may
be used. Machine learning algorithm may also be used. This may advantageously
allow a better
identification of locations infested with resistant weeds.
Optionally, the determination of one or a plurality of target areas for weed
management may
further comprise identifying S45 a single or a plurality of target areas with
at least one resistant
weed based on historical information of the plantation field. For example,
farmers may practice
precision techniques by utilizing weed infestation maps of previous seasons.
Weed stresses
that impact the crop are persistent. In general, if there was a weed present
in the last cultivation,
there probably be a weed in the next cultivation. The location of this weed
stress is known and
can be geo-located within e.g. inches. These stresses, including resistant
weeds, may be
identified and monitored for when and where to look to detect the next
emergence. Historical
data may include a weed map generated in the last cultivation, pre and during
Harvest
Burndown, or application sequences in autumn and spring. This may
advantageously allow the
13
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
application of secondary weed control technique only at locations prone to
weed infestation. For
example, herbicides may be applied to the plantation field either as pre-
emergent spray to soil
or incorporated at a depth, i.e. soil-injection.
Optionally, the determination of one or a plurality of target areas for weed
management may
further comprise identifying S46 a single or a plurality of target areas,
where the use of crop
protection products is not recommended, based on historical information of the
plantation field.
An application of a mechanical weed control technique is recommended in the
single or the
plurality of the identified target areas Historical data may include a weed
map generated in the
last cultivation, pre and during Harvest Burndown, or application sequences in
autumn and spring.
Farmers may also utilize weed infestation maps of previous seasons with areas
indicating that
the use of chemical control is not or less effective. Thus, in the present
season, the farmers can
use a mechanical control technique for a better weed management of these
areas.
.. Optionally, the determination of one or a plurality of target areas for
weed management may
further comprise determining S47, for each determined target area, a
respective secondary weed
control technique. The treatment control signal is indicative of the
respective determined
secondary weed control technique for each target area, which when transmitted
causes an
activation of at least one weed treatment device to apply, for each determined
target area, the
respective determined secondary weed control technique. The application of at
least one
secondary weed control technique comprises applying S51, for each determined
target area, the
respective determined secondary weed control technique. The secondary weed
control technique
may be determined based on historical information of the plantation field,
e.g. areas where a
mechanical control technique is recommended based on the weed map of previous
seasons. For
.. example, the secondary weed control technique may be determined based on
the determined
weed type and/or intensity. This may advantageously allow determining, for
each target area,
which weed control technique, e.g. chemical or mechanical, to be applied to.
This may also allow
determining what kind of herbicide to be used. This may advantageously offer
an improved weed
control at a local scale.
Optionally, the method may further comprise the step of generating S60 a field
record
comprising at least one of the following: i) the generated weed pressure map,
ii) the determined
geographical information of the single or the plurality of target areas with a
weed pressure
above a predefined threshold, iii) the determined weed type and/or the
determined weed density
.. of the single or the plurality of target areas, iv) the identified rows
and/or the identified density of
the single or the plurality of crops in the plantation field, v) the
identified single or the identified
plurality of target areas with at least one resistant weed, and vi) the
identified one or plurality of
14
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
target areas to be applied with a mechanical weed control technique. The field
report may be
used for online or offline precision farming.
Fig. 2 schematically shows an embodiment of a decision-support system 100 for
weed treatment.
The decision-support system 100 may be a server that provides a remote service
e.g. via internet,
to facilitate management of image data from many different plantation fields.
The server may
include an interface through which a user (e.g. a farmer) can manage the image
data and related
information. For example, the decision-support system 100 may interface with
users with
webpages served by the decision-support system to facilitate the management of
the image data
and related information. The related information may include e.g. one or more
target areas to be
treated, secondary weed control technique to be applied for these target
areas, configuration
information of one or more weed treatment devices. The configuration
information of the one or
more weed treatment devices may include, but not limited to, the start time of
the treatment, the
geographical information of the target areas, and the type of ingredient to be
used. These
information may be loaded onto the one or more weed treatment devices e.g. via
a network, to
enable the one or more weed treatment devices to perform a secondary weed
control on the
target areas. Alternatively, the decision-support system 100 may be a local
computing device,
such as a personal computer (PC).
In the following, the decision-support system 100 is discussed together with
Fig. 4, which
schematically shows a general set up of a plantation field 10.
The decision-support system 100 comprises a data interface 110, an image
recognition unit 120,
a weed pressure mapping unit 130, a determination unit 140, a controlling unit
150, and a
treatment control interface 160.
The data interface 110 may be a secure digital (SD) memory card interface, a
universal serial bus
(USB) interface, a Bluetooth interface, a wireless network interface, etc.
suitable to receive the
image data collected using ground, satellite or drone platforms. The image
data may comprise
spectral data or thermal data that can be used to discriminate weeds and crops
and preferably to
discriminate different types of weeds. The image data may also comprise GPS
data adapted for
providing guidance of the weed treatment devices to the target areas.
The image recognition unit 120 is configured to recognize items 12a, 12b, 12c
indicative of an
occurrence of a weed in image data of the plantation field 10 received from
the data interface
110. For example, Fig. 4 illustrates a plurality of items 12a, 12b, and 12c
indicative of an
occurrence of a weed. The items 12a, 12b, and 12c are illustrated in different
shapes for indicating
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
that they have different spectral or thermal responses in the image data and
therefore may
represent different types of weeds that require different treatment.
The weed pressure mapping unit 130 is configured to generate a weed pressure
map indicative
of a weed pressure in the plantation field based on the recognized items. For
example, the
plantation field may be divided into a plurality of grids in form of a
rectangular array of squares
or rectangles of equal size. The weed pressure may be evaluated for each grid
for generating a
weed pressure map for the plantation field.
The determination unit 140 is configured to determine one or a plurality of
target areas 14a, 14b,
14c for weed management in the generated weed pressure map, if the weed
pressure of the one
or the plurality of target areas extends a predetermined threshold. The
determination unit 140
may determine the target area or areas if the weed pressure value, e.g. in a
grid, is above a
predefined threshold, which may indicate the existence of weeds.
The controlling unit 150 is configured to generate a treatment control signal
indicative of the one
or the plurality of determined target areas and to output the treatment
control signal to the
treatment control interface, which when transmitted causes an activation of at
least one weed
treatment device to apply at least one secondary weed control technique to the
one or the plurality
of determined target areas. The controlling unit 150 may be a smartphone or a
mobile device to
be used by a farmer in the plantation field for regulating the activity of
weed treatment devices.
The controlling unit 150 may also be an integrated part of a computer decision-
support system
for regulating the activity of weed treatment devices.
Optionally, the determination unit 140 may be configured to determine, for
each determined target
area, a respective secondary weed control technique. The treatment control
signal is indicative
of the respective determined secondary weed control technique for each target
area, which when
transmitted causes an activation of at least one weed treatment device to
apply, for each
determined target area, the respective determined secondary weed control
technique. In an
example, the secondary weed control technique may be determined based on
historical
information of the plantation field, e.g. areas where a mechanical control
technique is
recommended based on the weed map of previous seasons. In a further example,
the secondary
weed control technique may be determined based on the actual information based
on the
collected image data, such as weed types and/or intensity.
16
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
Thus, the decision-support system may allow image processing, locating weeds,
noting the
intensity of infestation and mapping them as well as regulating the activity
of weed treatment
devices.
Fig. 3 schematically shows an embodiment of a weed treatment device 200 for
weed treatment
in the plantation field. In the following, the weed treatment device is
discussed together with Fig.
4.
The weed treatment device 200 comprises a treatment control interface 260 and
a treatment
arrangement 60 with one or a plurality of treatment units 61, 62, 63, 64. The
weed treatment
device 200 may be at least one selected from a mechanical weed treatment
device configured
to be steered by the controlling unit to apply a mechanical weed control
technique to the one or
the plurality of target areas, and a chemical weed treatment device configured
to apply a
second crop protection product to the one or the plurality of target areas.
For example, the weed
treatment device 200 may be e.g. ground robots with variable-rate applicators,
aerial sprayers,
or other variable-rate herbicide applicators. The treatment arrangement may be
a nozzle
arrangement comprising a plurality of nozzles as treatment unit. As a further
example, the weed
treatment device 200 may be a physical/mechanical treatment device with an
arrangement of
mechanical means as treatment units.
The treatment control interface 260 of the weed treatment device is
connectable to the treatment
control interface 160 of the decision-support system 100 as discussed in Fig.
2. This may be done
wirelessly, thus allowing a remote control of the weed treatment device 200
via the decision-
support system 100. The treatment control interface 260 is configured to
receive a treatment
control signal indicative of one or a plurality of target areas 14a, 14b, 14c
for weed treatment as
shown in Fig. 4.
The weed treatment device 200 is configured to activate respective ones of
treatment units 61,
62, 63, 64 of the treatment arrangement 60 to apply a secondary weed control
technique to the
one or the plurality of target areas 14a, 14b, 14c based on the received
treatment control signal.
Optionally, the treatment control signal is indicative of a respective
secondary weed control
technique for each target area 14a, 14b, 14c. The weed treatment device 200 is
configured to
apply the secondary weed control technique to a target area 14a, 14b, 14c
where the indicated
secondary weed control technique matches that of the weed treatment device.
For example, in
Fig. 4, multiple weed treatment devices 200a, 200b, 200c are deployed in the
plantation field 10
for weed treatment in the target areas 14a, 14b, 14c, each having a weed
control technique that
17
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
matches the identified secondary weed control technique of the respect target
areas to be
treated. For example, the weed treatment device 200a may be a mechanical weed
treatment
device for applying mechanical weed control to the target area 14a. The weed
treatment
devices 200b, 200c may be chemical weed treatment devices, which apply
different herbicides
to the target areas 14b,14c. Thus, the weeds in each target area may be
treated differently with
an appropriate weed control technique.
Thus, the weed treatment device, which is regulated by the decision-support
system, may allow
precision weed control techniques based on the weed infestation map.
Fig. 5 schematically shows an embodiment of a system 300 for weed treatment in
the plantation
field. The system comprises a sensor carrier 50, a decision-support system 100
as described
above and at least one weed treatment device as described above. The sensor
carrier 50, the
decision-support system 100 and the at least one weed treatment device 200 may
be
associated with a network. For example, the network may be the internet. The
network may
alternatively be any other type and number of networks. For example, the
network may be
implemented by several local area networks connected to a wide area network.
The network
may comprise any combination of wired networks, wireless networks, wide area
networks, local
area networks, etc. In some implementations, the decision-support system 100
may be a server
to provide a web service to facilitate weed management of a plantation field.
The user (e.g. a
farmer) may collect image data with a drone in his plantation field. He may
upload the image
data e.g. via the network, to the decision-support system 100 for analysis.
The decision-support
system 100 may output a treatment control signal comprising the configuration
information of
the weed treatment devices for activating these weed treatment device to
perform weed
management.
The sensor carrier 50 is configured to collect image data of the plantation
field after being
applied with a primary weed control technique. The sensor carrier 50 may be
e.g. ground
robots, satellite or drones fitted with visual, IR, N IR, and/or thermal
sensors for collecting the
image data.
The decision-support system 100 is configured to determine one or a plurality
areas for weed
control and preferably, for each determined target area, a respective
secondary weed control
technique based on the collected image data and to control the at least one
weed treatment
device 200 to apply at least one secondary weed control technique to the one
or the plurality of
target areas.
18
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
It has to be noted that embodiments of the invention are described with
reference to different
subject matters. In particular, some embodiments are described with reference
to method type
claims whereas other embodiments are described with reference to the device
type claims.
However, a person skilled in the art will gather from the above and the
following description that,
unless otherwise notified, in addition to any combination of features
belonging to one type of
subject matter also any combination between features relating to different
subject matters is
considered to be disclosed with this application. However, all features can be
combined providing
synergetic effects that are more than the simple summation of the features.
While the invention has been illustrated and described in detail in the
drawings and foregoing
description, such illustration and description are to be considered
illustrative or exemplary and
not restrictive. The invention is not limited to the disclosed embodiments.
Other variations to the
disclosed embodiments can be understood and effected by those skilled in the
art in practicing a
claimed invention, from a study of the drawings, the disclosure, and the
dependent claims.
In the claims, the word "comprising" does not exclude other elements or steps,
and the indefinite
article "a" or "an" does not exclude a plurality. A single processor or other
unit may fulfil the
functions of several items re-cited in the claims. The mere fact that certain
measures are re-cited
in mutually different dependent claims does not indicate that a combination of
these measures
cannot be used to advantage. Any reference signs in the claims should not be
construed as
limiting the scope.
19
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
Reference list
plantation field
12a item indicative of an occurrence of weeds
12b item indicative of an occurrence of weeds
5 12c item indicative of an occurrence of weeds
14a target area
14b target area
14c target area
10 50 sensor carrier
60 treatment arrangement
61 treatment unit
62 treatment unit
63 treatment unit
64 treatment unit
100 decision-support system
110 data interface
120 image recognition unit
130 weed pressure mapping unit
140 determination unit
150 controlling unit
160 treatment control interface
200 weed treatment device
200a weed treatment device
200b weed treatment device
200c weed treatment device
260 treatment control interface
300 system
S10 collecting image data
S20 recognizing items
S30 generating a weed pressure map
S40 determining one or a plurality of target areas
CA 03134436 2021-09-21
WO 2020/201046
PCT/EP2020/058646
S41 determining a geographical information
S42 recognizing a single or a plurality of weeds
S43 determining a weed type and/or a weed density
S44 identifying rows and/or a density of a single or a plurality of crops
S45 identifying a single or a plurality of target areas with resistant weeds
S46 identifying a single or a plurality of target areas, where the use of crop
protection products
is not recommended
S47 determining, for each determined target area, a respective secondary weed
control
technique
S50 applying at least one secondary weed control technique
S51 applying, for each determined target area, a respective secondary weed
control technique
S60 generating a field record
21
